URINARY STONES AND INTERVENTION QUALITY OF LIFE (USIQOL): DEVELOPMENT AND VALIDATION OF A NEW CORE UNIVERSAL PATIENT-REPORTED OUTCOME MEASURE FOR ...
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
EUF-1034; No. of Pages 8
E U RO P E A N U R O L O GY F O C U S X X X ( 2 019 ) X X X– X X X
available at www.sciencedirect.com
journal homepage: www.europeanurology.com/eufocus
Urinary Stones and Intervention Quality of Life (USIQoL):
Development and Validation of a New Core Universal Patient-
reported Outcome Measure for Urinary Calculi
Hrishikesh B. Joshi a,b,*, Hans Johnson c, Amelia Pietropaolo d, Aditya Raja a,b, Adrian D. Joyce e,
Bhaskar Somani d, Joe Philip c, Chandra Shekhar Biyani e, Tim Pickles a
a
School of Medicine, Cardiff University, Cardiff, UK; b Department of Urology, University Hospital of Wales, Heath Park, Cardiff, UK; c University of Bristol
Medical School, Bristol, UK; d University Hospital Southampton NHS Foundation Trust, Southampton, UK; e Leeds Teaching Hospitals NHS Trust, Leeds, UK
Article info
Associate Editor: Malte Rieken Keywords:
urinary calculi
health related quality of life
(HRQoL)
patient reported outcome
growing interest in the use of PROMs in routine HRQoL
* Corresponding author. Department of Urology, University Hospital of Wales, Heath Park, Cardiff
CF14 4XW, UK. Tel. +44 29 20745151.
E-mail address: hrishi.joshi@wales.nhs.uk (H.B. Joshi).
1. Introduction monitoring and medical audits [12]. A PROM can improve
the evidence base as long as the measure is appropriate and
Urolithiasis is a common condition with the prevalence of in accordance with international standards [13]. American
2–3% among the general population and 50% of patients Urological Association guidelines state that treatment deci-
likely to form further stones within 5 yr [1]. The prevalence sions about urinary calculi should incorporate patient pre-
rates reported range from 7% to 13% in North America, 5–9% ferences that are influenced by the HRQoL impact [14].
in Europe, and 1–5% in Asia [2]. The disease resulted in 550 Attempts have been made to measure the HRQoL of
000 emergency room visits in the USA in 2009 and more patients with urolithiasis. Generic measures have been used
than 30 800 hospital admissions in England in a year for this, but often fail to elaborate on the clinically relevant
[3,4]. Stone patients miss an average of 47.9 h of work domains [8]. In the recent past, PROMs specific to urolith-
per year, with additional hours lost due to ambulatory care iasis targeted at different subpopulations have been devel-
visits [5]. There are different options for managing urinary oped [15–17]. It is now recognised that modern psychomet-
calculi with expectant, medical, or interventional treat- ric methods based on Rasch measurement theory (RMT)
ments [6], which can be multistaged and carry different should be integral to the development of such measures
risks and success rates. Temporary interventions such as [18–20].
indwelling ureteric stents add to the patient burden [7]. Uro- Our hypothesis was that the subjective QoL impact of
lithiasis and its treatment(s) have an adverse effect on stone disease and interventions can be measured objec-
health-related quality of life (HRQoL) and can compromise tively using a valid and reliable PROM developed using
all areas of patient functioning [8–10]. modern methodology. Our aim was to develop a core PROM,
A patient-reported outcome measure (PROM) is a report incorporating RMT, to evaluate the impact of the entire
on a patient’s health condition that comes directly from the spectrum of upper-tract urinary calculi in a uniform way
patient [11]. In addition to their use in randomised con- and facilitate cross-comparison of the HRQoL impact of
trolled trials to assess treatment effectiveness, there is urolithiasis and interventions.
https://doi.org/10.1016/j.euf.2020.12.011
2405-4569/© 2020 European Association of Urology. Published by Elsevier B.V. All rights reserved.
Please cite this article in press as: Joshi HB, et al. Urinary Stones and Intervention Quality of Life (USIQoL): Development and
Validation of a New Core Universal Patient-reported Outcome Measure for Urinary Calculi. Eur Urol Focus (2021), https://doi.org/
10.1016/j.euf.2020.12.011
EUF-1034; No. of Pages 8
2 E U RO P E A N U RO L O GY F O C U S X X X ( 2 019 ) X X X– X X X
2. Patients and methods 2.3. Rasch analysis
We followed international PROM guidelines for the devel- Phase 4 assessed different properties of the USIQoL such as
opment and validation of the Urinary Stones and Interven- item and person locations, item fit (fit residuals and x2
tion Quality of Life (USIQoL) measure that would also statistics), Person Separation Index (PSI), response catego-
conform to the COSMIN (Consensus-based Standards for ries, and local dependence (Table 1) [24]. Misfitting items
the Selection of Health Measurement Instruments) check- were removed in an iterative manner, with removal of a
list [20]. The multicentre developmental process comprised single item at a time, after which the analyses were run
five stages (ethical approval: 17/WA/0195, no. 138478, again.
217163). Adult patients with urolithiasis covering all index In phase 5, in addition to the analyses in phase 4, we
stone categories, representative of routine practice, were assessed: (1) differential item functioning (DIF) for the traits
invited to participate. The participants included patients age (four groups), sex, stone site (kidney/ureter), type of
with renal or ureteric stones with or without treatment(s). intervention, presence or absence of symptoms, and history
The key steps are outlined in Fig. 1. of previous stones; (2) Smith’s test of unidimensionality
[25]; and (3) the optimal scale structure and logit-based
2.1. Development steps scoring.
2.1.1. Phases 1–3 2.4. Traditional analysis (internal consistency and validity)
The work in phases 1–3, with patient interviews involving
many stakeholders, produced a working conceptual frame- In phase 4, interitem and corrected item-total correlations
work and an initial long draft of the questionnaire were calculated. Correlations between scales (EuroQoL EQ-
[8,21]. After pretesting, the revised draft was administered 5D-5 L instrument, Short-Form-12 questionnaire, Hospital
in field test 1. Anxiety and Depression Scale, and Work Productivity and
Activity Impairment Questionnaire [expected 0.3–0.5])
2.1.2. Phase 4: field test 1 were assessed for criterion validity [26–29].
Field test 1 was undertaken to construct USIQoL scales and In phase 5, in addition to the analyses for phase 4, we
perform a preliminary psychometric evaluation in a large conducted tests of reliability (test-retest, patients with
sample to select the most appropriate items. stable disease completing the USIQoL twice, 24–72 h apart)
and validity (convergent) including both within- and
2.1.3. Phase 5: field test 2 between-scale testing and responsiveness to change (sub-
Field test 2 was undertaken to comprehensively evaluate group completing the USIQoL before and after interven-
the shortened version generated from phase 4 to produce tional treatments at an interval of 4–16 wk).
the final draft. Patients also completed existing generic
questionnaires and those with an indwelling stent com- 3. Results
pleted the Ureteric Stent Symptoms Questionnaire (USSQ)
[22]. 3.1. Item generation
2.2. Sample size considerations and statistical analysis A total of 62/77 invited patients (mean age 51 yr) and
30 family members participated in phases 1 and 2, gener-
The rule-of-thumb sample size recommendations for tradi- ating 106 themes and 10 broad headings. These were
tional analysis (10 subjects/item of the largest subscale mapped to a conceptual framework, with removal of dupli-
[18 items in the long draft; n = 180]) and Rasch analysis cations to create item sets [8,21]. A five-point rating scale
(n = 200 minimum and 400/500 maximum for four/five (ranging from “not at all” to “a lot”) was selected for the
class intervals) were followed for all assessments during initial draft (Fig. 1).
phases 4 and 5.
A combination of traditional and RMT assessments was 3.2. Pretesting
conducted using a sophisticated mathematical measure-
ment model [23]. SPSS 25 software (SPSS Inc., Chicago, IL, Forty patients evaluated USIQoL, with minor changes to the
USA) was used to perform traditional (eg, Spearman corre- items providing preliminary evidence of its face content
lations) analysis. Rasch analysis (polytomous extended validity and clinical suitability. A review by clinicians con-
response category, partial credit model) was performed firmed its completeness. The revised versions, with 60 items
using RUMM 2030 software. including treatment items, were drafted for the first field
test. This evaluated pain using different formats (frequency
of mild to unbearable pain, intensity of worst, day-to-day as
well as average pain; 10 items), physical and social health
including sex life (18 items), psychological health (six
Please cite this article in press as: Joshi HB, et al. Urinary Stones and Intervention Quality of Life (USIQoL): Development and
Validation of a New Core Universal Patient-reported Outcome Measure for Urinary Calculi. Eur Urol Focus (2021), https://doi.org/
10.1016/j.euf.2020.12.011
Table 1 –
Glossary of EUF-1034; No. of Pages 8
terms.
E U R O P E A N U R O L O GY F O C U S X X X ( 2 019 ) X X X– X X X 3
Generic patient-reported outcome measures
EQ-5D-3L: EuroQol 5-dimension, 3-level questionnaire. Descriptive system
for health-related quality of life states in adults; a preference-based measure
also used for economic appraisals to calculate quality-adjusted life years
SF-12: Medical Outcomes Study Short-Form-12 questionnaire. A 12-item
self-reported outcome measure assessing the impact of health on an
individual’s everyday life. Also used for preference-based utility (economic)
assessments.
Disease- and intervention-specific patient-reported outcome measures
USIQoL: Urinary Stones and Intervention Quality of Life questionnaire.
Quality-of-life measure specific for urinary stone (upper tract) disease and
interventions.
WISQoL: Wisconsin Stone Quality of Life questionnaire. Quality-of-life
measure specific for kidney stone disease.
USSQ: Ureteral Stent Symptoms Questionnaire. Intervention-specific
measure to assess the impact of ureteral stents on quality of life.
HADS: Hospital Anxiety and Depression Scale. Instrument used to measure
anxiety and depression in a general medical population of patients.
WPAI: Work Productivity and Activity Impairment Scale. Instrument used to
measure impairments in work.
Rasch measurement analysis terminology
(item = question, trait = patient/disease characteristics)
Logit range: For information on the scale to the sample target, a measure of
the match between the range of HRQoL (domains) measured with the
USIQoL and the range of HRQoL in the patient sample.
Targeting and Person Separation Index (PSI): Used to measure the reliability
of a scale. A questionnaire is perfectly targeted if the mean of the person is
the same as the mean of the items on the shared metric. PSI represents the
extent to which items distinguish between distinct levels of disease-specific
bother.
Item fit: The x2 statistic is used to confirm that the central property of item
invariance (the hierarchical ordering of the items) does not vary across the
trait measured. Fit residuals are used to assess differences between the
observed and expected data for each person and item. Fig. 1 – Steps in the development (phases 1–2) and evaluation (phases 3–5) of Urinary Stones
Ordered thresholds: Consistent use of the scale that corresponds to evidenceand Intervention Quality of Life (USIQoL).
that the response categories represent increasing levels of the construct HRQoL = health-related quality of life; EAU = European Association of Urology; BAUS = British
Association of Urological Surgeons; AUA = American Urological Association; PROM = patient-
being measured (the correct ordering of the response categories is reflected
reported outcome measure.
in successive thresholds). a
Construct definition: PROM development underpinned by the theory (conceptual base) and
Residual correlation: The extent to which each item is independent of the
others (helps to remove redundant questions).
items), work generic measures, range 0.3–0.8), demonstrating satisfac-
performance (eight items), and travel/holiday issues (three tory early item level validity.
items). Fourteen items addressed additional problems Using an iterative approach, misfitting and redundant
including treatments and help from health care providers items were removed to generate the revised versions for
and family members, and one item on global health. stone disease and interventions (20 and 24 items, respec-
tively). These included five scales for pain, social health (five
3.3. Field test 1: item reduction and scale development items each), physical health, psychological health (four
items each), and work (two items), with four treatment
During the first field test, 212/250 patients completed the items ready for field test 2.
questionnaires (Table 2). We evaluated psychometric prop-
erties, considering this to be a single scale, and seven 3.4. Field test 2
subscale formats.
Rasch analysis demonstrated important features of USI- In total, 369/390 patients participated in phase 5 (409 obser-
QoL, including limitations, requiring modifications. All vations, 61 patients completed >1 [pre- and post-treat-
scales indicated good to excellent reliability (PSI 0.62– ment] questionnaires), with 24/30 patients completing
0.89; Table 3). However, for almost all scales, more than the test-retest study (Table 2).
60% of the items had disordered thresholds (difficulty dis-
tinguishing between responses “quite a bit” and “very 3.4.1. Rasch analysis
much”) necessitating a change from five to four or two Rasch analysis demonstrated that most of the items in the
response categories. Each scale had items with significant scales mapped out continua of increasing bother (Table 3).
fit residuals (12–60%), and residual correlations (50–90%), The scales located items in a clinically sensible order with a
indicating item redundancy. good sample match. Deviations from model expectations
Traditional analysis showed that USIQoL is a reliable and were marginal.Items excluded were pain (life interference,
valid measure of the impact of stones on different domains. average and mild pain), social (sex, social life, and holiday),
Reliability was excellent (total scale, a = 0.9, subscales, a = psychological (worry about kidney failing), and treatment
0.6–0.9). The corrected item total (0.3–0.8) and interitem (diet and device). The two treatment items (medication,
(0.4–0.9) correlations were satisfactory. Preliminary analy- water intake) were combined with the social scale. This
ses of criterion validity were as expected (correlations with transformed the USIQoL into a final 15-item measure.
Please cite this article in press as: Joshi HB, et al. Urinary Stones and Intervention Quality of Life (USIQoL): Development and
Validation of a New Core Universal Patient-reported Outcome Measure for Urinary Calculi. Eur Urol Focus (2021), https://doi.org/
10.1016/j.euf.2020.12.011EUF-1034; No. of Pages 8
4 E U RO P E A N U RO L O GY F O C U S X X X ( 2 019 ) X X X– X X X
Table 2 – Participant characteristics.
Characteristic Field test 1 Field test 2
a
(n = 212/250) (n = 345/360)
Mean/median age (yr) 52.5/52 53.8/56
Age range (yr) 19–89 18–90
Age group, n (%)
16–40 yr 64 (30) 98 (29)
41–64 yr 92 (43) 166 (48)
65–80 yr 50 (24) 74 (21)
>80 yr 6 (3) 7 (2)
Sex, n (%)
Male 135 (64) 241 (69)
Female 77 (36) 103 (31)
Not recorded 1
Bothersome pain, n (%)
Yes 127 (65) 204 (60)
No 73 (35) 138 (40)
Not recorded 12 3
Site of stone, n (%)
Kidney 147 (68) 225 (63)
Ureter 62 (32) 115 (35)
Not recorded 3 5
Previous stones, n (%)
Yes 116 (56) 176 (52)
No 84 (44) 162 (48)
Not recorded 12 7
Paid employment (n)
Yes 125 230
No 81 115
Not recorded 6
Current treatment(s), n (%)
Medical (metabolic disorder) 13 (6) 28 (8.1)
Shockwave lithotripsy 73 (34.4) 122 (35.3)
Surgical interventions (eg, URS/PCNL) 37 (18.3) 79 (22.8)
Observation ( short-term medical treatment, eg, tamsulosin, analgesics) 88 (41.3) 117 (33.8)
Stent in situ 16 (7.5) 33 (9.6)
URS = ureteroscopy; PCNL = percutaneous nephrolithotomy.
a
Test-retest data not included.
Table 3 – Psychometric testing results from the Rasch analysis for field test 1 and field test 2.
a a b d
USIQoL scale Items (mean) Persons (mean) PSI Fit statistics Disordered Residual correlation U
Locations FR Locations FR Items with FR outside 2.5 SD (n)Items with x2thresholds cItems with r score range > +0.3 (n)
(logit range) (logit range) p > 0.001 (n)
Field test 1 (n = 212)
Total (59 items) 0.0 0.08 0.27 0.450.8814 10 55 56 N
Pain (10 items) 0.0 0.94 0.36 0.690.844 2 6 7 N
Physical + social (18 items) 0.0 0.46 0.08 0.600.894 2 16 12 N
Psychological (6 items) 0.0 0.88 0.34 0.980.891 1 0 3 N
Work (8 items) 0.0 0.85 1.10 0.620.817 5 8 6 N
Travel (3 items) 0.0 0.87 0.66 0.500.620 2 3 2 N
Field test 2 (n = 345 with 409 observations)
Pain and Physical Health (6)0.00 0.880.08 0.610.720 ( 2.0 to +1.3) 0 0 0 ( 0.3 to+0.07) Y
Psychosocial Health (7) 0.00 0.580.31 0.550.700 ( 1.0 to +0.3) 0 0 0 ( 0.4 to +0.06) Y
Work (2) 0.00 1.14 2.58 1.050.830 ( 1.3 to 0.9) 0 0 0 ( 0.98) Y
PSI = person separation index (measures the reliability of the scale; 0.7 is adequate); FR = fit residual; SD = standard deviation.
a
Logit range: for information on the scale to the sample target, the match between the range of HRQoL measured with the USIQoL and the range of HRQoL in the
patient sample.
b
Item fit measured in terms of (1) fit residual (expected to lie between a mean of 0 and 2.5 SD) and (b) x2 statistic (should be less than the Bonferroni-corrected
significance level).
c
Disordered thresholds: response categories not working as intended (measured using item response curves and threshold maps).
d
Residual correlation: the extent to which each item is independent of the others (should beEUF-1034; No. of Pages 8
E U R O P E A N U R O L O GY F O C U S X X X ( 2 019 ) X X X– X X X 5
3.4.2. Revised scaling 3.4.3. DIF and unidimensionality
Items had superior fits when the five-scale structure was We evaluated all 15 questions and three scales against
changed to a three-scale form, combining pain and physical different patient subpopulations (Supplementary Table 1).
health domains (PPH; six items), psychological and social This confirmed good performance across traits. All three
health domains (PSH; seven items), and the work domain scales were unidimensional.
(two items). Fig. 2 demonstrates satisfactory item-threshold
distribution maps for the subscales. 3.4.4. Traditional analysis
Traditional analysis confirmed that all three USIQoL scales
are reliable and valid measures for assessment of important
Fig. 2 – Person-item threshold distribution map for (A) Pain and Physical Health Scale (grouping set to interval length of 0.20, making 55 groups), (B)
Psychosocial Health Scale (grouping set to interval length of 0.20, making 45 groups), and (C) work (grouping set to interval length of 0.20, making
70 groups).
SD = standard deviation.
Please cite this article in press as: Joshi HB, et al. Urinary Stones and Intervention Quality of Life (USIQoL): Development and
Validation of a New Core Universal Patient-reported Outcome Measure for Urinary Calculi. Eur Urol Focus (2021), https://doi.org/
10.1016/j.euf.2020.12.011EUF-1034; No. of Pages 8
6 E U RO P E A N U RO L O GY F O C U S X X X ( 2 019 ) X X X– X X X
domains across patient groups. Corrected item-total and 4. Discussion
interitem correlations supported the hypothesis that items
within scales measured a common underlying construct The recurrent nature of urolithiasis and ensuing interven-
with good reliability. Test-retest correlations were excellent tions can result in a cumulative negative HRQoL impact. The
(0.81–0.92), indicating good scale stability (Table 4). impact can be assessed using PROMs, but this involves
Criterion validity was tested extensively and hypothe- measurement challenges. Generic measures fail to capture
sised correlations between scores for USIQoL scales and this impact comprehensively. Hence, urolithiasis-specific
existing generic and domain-specific measures were con- PROMs have recently been developed. The Wisconsin Stone
sistent. We showed that there was very good correlation Quality of Life (WISQoL, 28 items) was the first measure
with the relevant domains between the USIQoL and USSQ developed to assess the impact of stable urolithiasis and
scales. The USIQoL was responsive to change, as shown by a medical therapies [15–17]. It has undergone linguistic vali-
significant positive effect for all scale scores after dations, with wide applications in different studies.
intervention. It is well recognised that measures that comply with
modern psychometric methods based on item response
3.4.5. Final USIQoL measure and scoring theory (RMT) are of higher quality [20]. In this respect,
The final USIQoL (three scales and 15 items; Supplementary the development of recent PROMs had a focus on specific
Table 2) is intended for self-administration, whereby subgroups and involved only traditional methods. These do
patients rate the amount of bother attributed on a 4-point not cover key criteria in the COSMIN guidelines (content
scale (1 = not at all, 2 = a little, 3 = quite a bit, and 4 = a lot). development, sample size, use of RMT, unidimensionality).
The disease and intervention versions are similar and differ The new USIQoL is the first PROM to capture the HRQoL
only in the title time frame (since your “current stone impact of urolithiasis (acute and chronic) and interventions.
problems” or “current or most recent stone treatment”) It was developed using a combination of classical and RMT
to make the versions psychometrically valid. Scale scores approaches, with very few such measures in urology. In the
are generated by summing items and transferring to a 0– Rasch model, the probability of a specified response (right/
100 (logit) scale, with high scores indicating greater patient wrong answer) is modelled as a function of person and item
bother. parameters. This is a unique mathematical modelling
approach based on a latent trait for which item values
Table 4 – Summary of traditional psychometric analysis: field test 2.
Test criterion USIQoL scales
PPH (6 items) PSH (revised 7-item scale; n = 156) Work performance (2 items)
Cronbach’s a 0.82 0.75 0.94
Interitem correlation range 0.29–0.56 0.11–0.59 0.89
Item total correlation range 0.51–0.62 0.0.35–0.58 NA
Test-retest (n = 24) 0.91 0.83 0.80
Construct validity (correlation coefficient)
EQ-5D-3 L (n = 346)
Utility-total 0.41 0.36 0.02
Pain/discomfort 0.49 0.39 0.12
Mobility 0.26 0.24 0.11
Usual activities 0.40 0.42 0.02
Anxiety/depression 0.28 0.51 0.12
EQ-5D-Thermometer 0.39 0.45 0.01
SF-12
Physical health (PCS, n = 296) 0.53 0.54 0.00
Mental health (MCS, n = 302) 0.37 0.46 0.15
WPAI (n = 67) 0.46 0.62 0.7
HADS
Anxiety (n = 166) 0.47 0.52 0.08
Depression (n = 163) 0.54 0.52 0.00
USSQ
Pain (n = 14) 0.71 0.30 NA
Urinary symptoms (n = 14) 0. 84 0.56 NA
General health (n = 14) 0.62 0.87 NA
PPTS effect size (n = 57) a 0.6 0.123 0.35
PPH = Pain and Physical Health; PSH = Psychosocial Health; EQ-5D-3L: EuroQol 5-dimension, 3-level questionnaire; SF-12 = Medical Outcomes Study short-
form questionnaire; PCS = physical component summary; MCS = mental component summary; WPAI = Work Productivity and Activity Impairment Scale; HADS
= Hospital Anxiety and Depression Scale; USSQ = Ureteral Stent Symptoms Questionnaire; PPTS = pre–post-treatment scale; NA = not applicable.
a
We assessed the responsiveness of USIQoL by calculating effect sizes in a comparison of pre-intervention and post-intervention scale scores. We expected these
to be positive, confirming a post-treatment improvement leading to reduced bother, but did not hypothesise a magnitude given the relatively smaller sample and
the first application of the scales.
Please cite this article in press as: Joshi HB, et al. Urinary Stones and Intervention Quality of Life (USIQoL): Development and
Validation of a New Core Universal Patient-reported Outcome Measure for Urinary Calculi. Eur Urol Focus (2021), https://doi.org/
10.1016/j.euf.2020.12.011EUF-1034; No. of Pages 8
E U R O P E A N U R O L O GY F O C U S X X X ( 2 019 ) X X X– X X X 7
are calibrated, and person abilities are measured on a and its wider application would need linguistic and cross-
shared continuum that accounts for the latent trait. This cultural validation. Its application with existing measures,
provides an internally valid measure that is independent of such as WISQoL, could help in capturing a broader picture.
the particular sample; the findings for the sample can be Further USIQoL application, including in daily practice, will
extrapolated to the population for measurement of clini- investigate scale sensitivity to develop clinically relevant
cally meaningful differences [30,31]. thresholds. There is scope for adaptations to undertake
The final 15-item selection for USIQoL was based on economic appraisals and compare emergent treatments
appraisals of the analyses against clinical relevance and and service evaluations that would guide patient-centred
measurement criteria. The psychometric evaluation care.
showed that all three scales satisfy criteria for acceptability,
validity, and reliability. The logit scoring for each scale offers 5. Conclusions
different scores that allow clearer identification of the
impact across different domains. This would help in future In conclusion, USIQoL is a new three-scale, 15-item, single-
comparative studies and sample size calculations. page self-report instrument that measures the HRQoL
The results from traditional validity assessments alone impact of stone disease and interventions. It has been
suggested that the long draft of the USIQoL satisfied most of developed using modern psychometric methods. It is fit
the criteria, until RMT demonstrated many targeting pro- for valid and reliable comparisons at the micro level
blems (disordered responses, item redundancies). This (patients) and meso level (treatment groups, institutions).
highlighted the value of RMT in conducting item-level We expect USIQoL to serve as a core PROM for studies
analyses that guide precise item selection and rectify pro- looking at and comparing the effectiveness of treatments,
blems with scales. Our analysis demonstrated that the five- observational strategies, and quality of care, as well as an
stage mixed-methods approach was important because of adjunct to medical audits. This PROM is expected to
the complex assessments involved. improve the evidence base and help toward better patient
communication and shared decision-making.
4.1. Strengths and limitations Author contributions: Hrishikesh B. Joshi had full access
to all the data in the study and takes responsibility for the
Many aspects important to different stakeholders were integrity of the data and the accuracy of the data analysis.
considered for the conceptual framework and subsequent Study concept and design: Joshi.
steps. Apart from construction of the necessary items and Acquisition of data: Johnson, Pietropaolo, Raja, Joshi.
scales based on the key themes, we carefully evaluated Analysis and interpretation of data: Joshi, Raja, Pickles.
whether there was a need for separate instruments for Drafting of the manuscript: Joshi, Biyani, Somani, Philip,
renal and ureteric stones, as well as the disease and inter- Pickles.
ventions. We also looked at the demonstrable applicability Critical revision of the manuscript for important intellectual
of the PROM and the uniformity of the performance across content: Joshi, Joyce, Biyani, Pickles.
the entire disease spectrum. Our work indicated that QoL in Statistical analysis: Joshi, Pickles.
relation to different sites, diseases, and treatments is inter- Obtaining funding: Joshi.
linked and separate measures can pose psychometric diffi- Administrative, technical, or material support: Biyani,
culties. The USIQoL development phases demonstrated that Joshi, Joyce, Pickles.
formulation of a single integrated PROM gave a better model Supervision: Joshi, Biyani, Somani, Philip.
of item fit and performed well across patient, disease, and Other: None.
intervention groups, making USIQoL an appropriate core Financial disclosures: Hrishikesh B. Joshi certifies that
instrument. all conflicts of interest, including specific financial interests
All three scales of the USIQoL demonstrated very good and relationships and affiliations relevant to the subject
performance with proven unidimensionality. It was matter or materials discussed in the manuscript (eg,
observed that pain, along with physical symptoms, which employment/affiliation, grants or funding, consultancies,
drive most of the clinical assessments, have more visible honoraria, stock ownership or options, expert testimony,
impact. The domain of pain, which is the most complex to royalties, or patents filed, received, or pending), are the
assess, was tested extensively before finalising its appro- following: None.
priate format for inclusion. Similarly, issues regarding work Funding/Support and role of the sponsor: This work was
are important to all stakeholders. The psychosocial scale is supported by funding from the Urology Foundation. The
likely to be a good indicator of issues not evaluated rou- sponsor played a role in the design and conduct of the study.
tinely and the longer-term impact of the condition, which Acknowledgments: We thank the Urology Foundation
could drive treatment choices. USIQoL captures all these for partial funding. We are grateful to Prof. K. Sarica, Dr. F.
dimensions well, with the results quantified using modern Sangueodolce (European Association of Urology), Mr. A.
psychometric techniques. USIQoL can also help in reliable Dickinson and BAUS Endourology Committee Members,
combined HRQoL evaluation for stent subgroups. Prof. M. Monga, and Dr. Sri Sivalingam (American Urological
There are certain limitations of the study and future Association). We also thank Prof. S. Salek for help with the
work would help to address these. USIQoL was developed study design.
and validated in the English-speaking population of the UK
Please cite this article in press as: Joshi HB, et al. Urinary Stones and Intervention Quality of Life (USIQoL): Development and
Validation of a New Core Universal Patient-reported Outcome Measure for Urinary Calculi. Eur Urol Focus (2021), https://doi.org/
10.1016/j.euf.2020.12.011EUF-1034; No. of Pages 8
8 E U RO P E A N U RO L O GY F O C U S X X X ( 2 019 ) X X X– X X X
[15] Penniston KL, Nakada SY. Development of an instrument to assess
Appendix A. Supplementary data the health related quality of life of kidney stone formers. J Urol
2013;189:921–30.
[16] Tran Mgb, Sut Mk, Collie J, et al. Development of a disease-specific
Supplementary material related to this article can be
ureteral calculus patient reported outcome measurement instru-
found, in the online version, at doi:https://doi.org/10.
ment. J Endourol 2018;32:548–58.
1016/j.euf.2020.12.011.
[17] Ragab M, Baldin N, Collie J, et al. Qualitative exploration of the renal
stone patients’ experience and development of the renal stone-
specific patient-reported outcome measure. BJU Int 2020;125:123–
References 32.
[18] Lord FM. Applications of item response theory to practical testing
[1] Pearle MS, Calhoun EA, Curhan G. Urologic diseases in America
problems. Hillsdale. NJ: Erlbaum; 1980.
project: urolithiasis. J Urol 2005;173:848.
[19] Rasch G. Probabilistic models for some intelligence and attainment
[2] Sorokin I, Mamoulakis C, Miyazawa K, Rodgers A, Talati J, Lotan Y.
tests (expanded edition). Chicago, IL: University of Chicago Press;
Epidemiology of stone disease across the world. World J Urol
1980.
2017;35:1301–20.
[20] Terwee CB, Prinsen CAC, Chiarotto A, et al. COSMIN methodology for
[3] National Health Service. Hospital episodes statistics data. Vol.
evaluating the content validity of patient-reported outcome mea-
2014. NHS. 2015.
sures: a Delphi study. Qual Life Res 2018;27:1159–70.
[4] Bultitude M, Rees J. Management of renal colic. BMJ 2012;345:
[21] Raja A, Wood F, Joshi HB. The impact of urinary stone disease and
e5499.
their treatment on patients’ quality of life: a qualitative study.
[5] Saigal CS, Joyce G, Timilsina AR, et al. Direct and indirect costs of
Urolithiasis 2020;48:227–34.
nephrolithiasis in an employed population: opportunity for disease
[22] Rasch G. Probabilistic models for some intelligence and attainment
management? Kidney Int 2005;68:1808–14.
tests. Copenhagen, Denmark: Danish Institute for Educational
[6] Türk C, Neisius A, Petrik A, Seitz C, Skolarikos A, Thomas K. EAU
Research; 1960.
guidelines: urolithiasis. Arnhem, The Netherlands: European Asso-
[23] Smith Jr EV. Metric development and score reporting in Rasch
ciation of Urology; 2020. http://uroweb.org/guideline/urolithiasis
measurement. J Appl Meas 2000;1:303–26.
[7] Joshi HB, Stainthorpe A, McDonagh RP, et al. Indwelling ureteral
[24] The EuroQol Group. EuroQoL—a new facility for the measurement
stents: evaluation of symptoms, quality of life and utility. J Urol
of health related quality of life. Health Policy (New York)
2003;169:1065–9.
1990;16:199.
[8] Raja A, Hekmati Z, Joshi Hb. How do urinary calculi influence health-
[25] Gandek B, Ware JE, Aaronson NK, et al. Cross-validation of item
related quality of life and patient treatment preference: a system-
selection and scoring for the SF-12 health survey in nine countries:
atic review. J Endourol 2016;30:727–43.
results from the IQOLA project. J Clin Epidemiol 1998;51:1171–8.
[9] Penniston KL, Nakada SY. Treatment expectations and health-
[26] Zigmond AS, Snaith RP. The hospital anxiety and depression scale.
related quality of life in stone formers. Curr Opin Urol
Acta Psychiatrica Scand 1983;67:361–70.
2016;26:50–5.
[27] Reilly MC, Zbrozek AS, Dukes EM. The validity and reproducibility of
[10] Patel N, Brown RD, Sarkissian C, De S, Monga M. Quality of life and
a work productivity and activity impairment instrument. Pharma-
urolithiasis: the patient - reported outcomes measurement infor-
coeconomics 1993;4:353–65.
mation system (PROMIS). Int Braz J Urol 2017;43:880–6.
[28] Joshi HB, Newns N, Stainthorpe A, MacDonagh RP, Keeley Jr FX,
[11] US Food and Drug Administration. Patient reported outcome mea-
Timoney AG. Ureteral stent symptom questionnaire: development
sures: use in medical product development to support labelling
and validation of a multidimensional quality of life measure. J Urol
claims. Guidance for Industry. Rockville, MD: US FDA; 2009.
2003;169:1060–4.
[12] Department of Health. Equity and excellence: liberating the NHS.
[29] Smith Jr EV. Detecting and evaluating the impact of multidimen-
London, UK: The Stationery Office; 2010.
sionality using item fit statistics and principal component analysis
[13] Aaronson N, Alonso J, Burnam A, et al. Assessing health status and
of residuals. J Appl Meas 2002;3:205–31.
quality of life instruments: attributes and review criteria. Qual Res
[30] Granger CV. Rasch analysis is important to understand and use for
2002;11:193–205.
measurement. Rasch Meas Trans 2007;21:1122–3.
[14] Assimos D, Krambeck A, Miller NL, et al. Surgical management of
[31] Hobart J, Cano S. Improving the evaluation of therapeutic interven-
stones: American Urological Association/Endourological Society
tions in multiple sclerosis: the role of new psychometric methods.
guideline, part I. J Urol 2016;196:1153–60.
Health Technol Assess 2009;13(12):1–200.
Please cite this article in press as: Joshi HB, et al. Urinary Stones and Intervention Quality of Life (USIQoL): Development and
Validation of a New Core Universal Patient-reported Outcome Measure for Urinary Calculi. Eur Urol Focus (2021), https://doi.org/
10.1016/j.euf.2020.12.011You can also read
